In response to:

The Pill of Pills from the June 9, 1994 issue

To the Editors:

I would be happy to have Listening to Prozac judged by Sherwin Nuland’s criteria for evaluating clinical evidence: meticulous observation, even-handed review of the literature, and attention to detail. Where we differ—beyond what it means to be even-handed in a review—is in our interpretation of his demand that doctors not “speculate beyond what is justified by the accumulated data and its supportable implications.” I suspect Nuland means that doctors must speculate only in private.

Researchers speculate constantly. Alternately profound and fanciful speculation determines the direction of pharmacologic inquiry. Whenever I speak to researchers, they engage in variously substantiated speculation about the functions of biological systems. Repeatedly, they say what I report in the book, that certain serotonin-based brain systems likely have to do with modulation of emotional function, influencing a person’s overall sense of well-being or status.

This speculation is well known to clinicians, and it affects clinical judgment, especially in the sorts of cases I discuss, where the diagnosis is murky and the indication for drug treatment controversial. The issue is not whether speculation goes on, but whether it should remain hidden within the medical establishment. Put differently:The indications for psychopharmacologic drug use are changing, via thousands of daily decisions made in thousands of doctors’ offices. What Listening to Prozac does is to lay out the basis for that change and to bring the attendant controversies into the public arena.

The form of Nuland’s review betrays his agenda. He goes on tediously establishing his bona fides as an old-style doctor (trained by the ostensibly witty Dr. Bean, familiar with the story of serotonin, etc., etc.)—and then fails to illustrate his attack with a single specific. The punch line is his claim that “only a reader familiar with the specialized literature” can understand his objections to the way I draw conclusions from research. Beyond insulting your readership, this “I could if I would” approach constitutes the lowest form of criticism; no informative discussion can arise, only measurement of the relative whiteness of coats.

As regards the weight of authority, Listening to Prozac has received numerous highly favorable reviews by colleagues (though I am unaware of other reviews by surgeons)in both the lay and professional press—including rave reviews, calling the book a landmark, in subspecialty journals—and has been adopted in the curricula of courses in pharmacology, psychiatry, and ethics. The central ideas in Listening to Prozac were presented to psychiatrists in my monthly column in the Psychiatric Times; for the most part, they have wide acceptance.

Beyond ex cathedra pronouncements, Dr. Nuland’s review contains a single substantive criticism, concerning the frequency of personality “transformation” in patients taking Prozac. Dr. Nuland quotes me correctly as writing that a “substantial minority” of patients experience this effect and counters by saying that the highest estimate he can find is 10 percent. Before Listening to Prozac was published, when Psychology Today was considering whether to excerpt, the editors consulted the late Daniel X. Freedman, editor of the AMA’s psychiatric research journal and a worrier about public enthusiasms for drugs. Dr. Freedman estimated that a transformative response occurred in 20 percent of users. Five million Americans had then taken Prozac; doing the math, the editors chose to run the selection.

These percentage estimates are next to meaningless, because transformation lacks an operational definition and because the figures lack reference to populations:10 (or 20) percent of whom?And if 10 percent are “transformed,” how many more people “merely” experience heightened confidence or lowered sensitivity to rejection? In the book, I ask whether the particular population I work with—relatively healthy, high-functioning people who have been in extensive psychotherapy—may not be the best medication responders, for a variety of reasons, including their ability to amplify the social effects of small changes in feeling states.

The argument of the book does not depend on a high frequency of transformation. Dr. Nuland concedes, “Of course, Prozac is capable of transformations…” Whence “of course?” This effect, though observed, had not been acknowledged in print before I wrote the professional essays that led to Listening to Prozac. And ought not clinicians reflect on what they see that is new or striking?That such apparently autobiographical traits as self-esteem even occasionally respond to medication seems a fit subject for public consideration.

My method is to make general readers part of an ongoing professional dialogue, familiarizing them with a variety of perspectives. Yes, I draw on diverse sorts of evidence to make plausible my thesis that antidepressants can affect personality traits even in people who do not suffer overt mental illness. But I take care to expose the reader to alternative viewpoints:that such medications affect only depressed mood, that when antidepressants appear to alter personality they are merely inducing hypomania, and so forth. Often I point to research that predicts something different from what I observe, for instance, evidence that Prozac should not, on theoretical grounds, influence a given trait, because the function in question is thought to be governed by brains systems that Prozac does not affect directly. Far from retailing “the magical prospect of being reborn with one’s defects of perspective purified,” I indicate that the traits of Prozac affects may constitute normal variants (or even moral virtues) whose alteration entails cause for concern.


The remainder of Dr. Nuland’s attack is ad hominem. He depicts me as a naive clinician who has “made himself into something like a [psychopharmacologic] messiah.” In truth, I am a psychiatrist previously known for his work in psychotherapy; in evaluating Prozac I unashamedly use the psychotherapist’s method, looking carefully at selected cases and examining their implications in the light of research. This approach is time-honored in psychiatry—distinct from, but not superseded by, controlled clinical trials, which in the mental health domain have problems of their own.

Though not a pharmacologist, I am aware of the field’s scope, methods, and history. After training at Harvard and Yale, I headed a Public Health Service division that had oversight for the entire Federal research portfolio in mental health and substance abuse. I am well aware of the history of enthusiasms for prescribed drugs; in the early 1980s I developed a report for the Surgeon General on women and the misuse of prescribed psychotropics. I have published peer-reviewed monographs on prescribing practices.

Most off target is Dr. Nuland’s repeated assertion that I am an evangelist for Prozac and for simple-minded psychobiology. At the heart of Listening to Prozac is a critique of contemporary biological materialism. I repeatedly cite situations where people overgeneralize from limited evidence concerning physiological underpinnings of personality. And so numerous are my concerns over the misuse, both clinical and ideological, to which Prozac can be put that a previous critic suggested they might form the basis for a second book titled “Worrying About Prozac.” There is, I hope, a saving element of irony in the title Listening to Prozac; ultimately drugs do not talk. In the end, what we listen to is our own extrapolations and projections.

I wonder whether Dr. Nuland is responding to the book or, rather, to the book’s popularity. As a book becomes popular, it became easy to mistake its genre.

Listening to Prozac is not self-help. It is not advocacy for a miracle cure. Writing the book, I had in mind a distinguished contemporary non-fiction genre, represented by John McPhee’s essays on frozen orange juice, Tracy Kidder’s on the Apollo computer, Witold Rybczynski’s on boat-building, Verlyn Klinkenborg’s on haying. Each author examines a particular technology and worries it until it resonates. To see the large in the small, to allow a product to illuminate the culture—these are the modern essayist’s goals. By mistaking the book’s genre—by taking curiosity for advocacy—Dr. Nuland creates a false need to defend the medical profession.

I am not ordinarily a doggerelist, but the example of the sainted Dr. Bean moves me to summarize:

The pharmacologic Messiah
Is a role to which I don’t aspire.
More than sing Prozac’s praises
I limn issues it raises.
(Sherwin, please doff the doctor’s attire.)

Peter D. Kramer
Providence, Rhode Island

Sherwin B Nuland replies:

I’m pleased to see Peter Kramer’s spirited defense of his book, but it does make me wonder why he has chosen to further expose the thinking patterns that have already led him so far astray.

If the author of Listening to Prozac was previously unaware of it, let him herewith be informed that responsible clinicians and researchers most definitely do not use scattered preliminary observations to make unjustified assertions in the pages of trade books aimed at a wide variety of general readers. Though it is true that “Researchers speculate constantly,” I’ve rarely heard of one who does his speculating for the millions, on TV and in newspapers. Limn Schmimn—this is not called science, it’s called hype, and it’s well orchestrated by professional publicists.

Dr. Kramer would have us believe that his is a form of speculation that “is well known to researchers.” Neither in its justification nor in the venues of its publications and promotion is that so, and it is hard to believe that a writer who claims familiarity with the basic ground rules of scientific reasoning is not aware of it. Many of his statements about Prozac are nothing more than a series of fanciful ruminations, mantled in the adornments of quasi-scientific logic and written in such a way that they cannot but mislead the unwary general reader.

In his letter, Dr. Kramer compounds the mischief by parading his academic accomplishments, and referring to the “numerous highly favorable” and even “rave” reviews his book has received. Apparently, that panoply of laurels is meant to convince us of the validity of his arguments, since he knows they can’t stand on their own. As he points out, however, under such a barrage, “no informative discussion can arise, only measurement of the relative whiteness of coats.” His credentialing cannot sidestep the issue, which is the credibility of his statements and the consequences of noising them throughout the land.


In fact, Dr. Kramer has so bleached his own coat that it separates him entirely (his grandiose comparisons notwithstanding) from the company of such as John McPhee, Tracy Kidder, and Verlyn Klinkenborg, whose justifiably acclaimed pieces work so well specifically because they were crafted by fascinated amateurs.

It seems odd that Peter Kramer should invoke the late Daniel X. Freedman to bolster his claims. Were Dr. Freedman alive to defend himself, this former president of the American Psychiatric Association (and a pioneer clinical psychopharmacologist)would remind us of his August 8, 1993, New York Times Book Review critique of Listening to Prozac, in which he said, “Readers should know that the number of patients who experience startling personality changes is rather small. Indeed it will be news to many of the millions of patients who take it for depression, obsessive-compulsive disorders, obesity, anxiety and eating disorders that this drug can cause dramatic changes in temperament.” Further on, Dr. Freedman writes that Listening to Prozac “does not provide reliable directions for those in troubled quest of self or those seeking a clear introduction to the brain sciences and pharmacology.” And then, “To fully grasp these effects [the biological accompaniments of psychological changes], Dr. Kramer would need a better understanding of science.”

Dr. Freedman does not burden his review with examples. Is his too, “the lowest form of criticism” or is it just one of those remarkably unfavorable reviews that Dr. Kramer chooses not to mention? Incidentally, the Freedman review refers to the Psychiatric Times as a “trade newspaper,” which is exactly what it is. Provided gratis to 40,000 psychiatrists, and published by an entrepreneur, it bears no resemblances to a medical journal.

By the way, I have never been near the University of Iowa, nor did I know Dr. Bean personally. For physicians of the last several generations, he has been an examplar of the kind of thinking that all professors of medicine taught students. It’s a shame that Dr. Kramer has never heard of him, especially because he died only five years ago. It’s not quite like knowing Lewis Thomas, but pretty close, at least for a doctor.

This Issue

July 14, 1994